Dear FSL group,
As you may know there is considerable debate regarding global signal regression in resting-state fMRI analysis. Though several methods have been proposed for removing physiological noise out of the data instead of global signal regression, there is no standard method for doing so. Hence, I am wondering what the preferred strategy of the FSL group would be when conducting a seed-based resting-state fMRI analysis?
Additionally, is it correct that global signal regression would only increase the probability of artifactual anticorrelations (negative functional connectivity), while having no adverse effects on, or even improve, positive correlations?
Thanks for your time and effort!
Best,
Mauricio Delgado
|